miller_06_offprint.pdf

7/27/2019 Miller_06_offprint.pdf

http://slidepdf.com/reader/full/miller06offprintpdf 1/13

An appreciation of others’ thoughts, feelings, knowl-edge, and wishes, or a ‘‘theory of mind,’’ is essentialfor competent communication. Without a theory of

mind, I may give you too much information, or too little;I may hurt your feelings, confuse you, or bore you. With atheory of mind, I can judge what you need and want to know.In typical development, theory of mind is so closely coupledwiththe development of communication and languagethat weoften do not recognize their interdependence. In children withdevelopmental disorders,however, deficits in languageand/or theory of mind may draw closer attention to the relationshipsbetween them. This tutorial will describe how language andtheory of mind development are related in typical and atypical

development, and it will suggest ways that clinicians mayapply this knowledge in working with children with languagedisorders of various etiologies.

This tutorial presents selected theoretical and empiricalresults that will help to sketch the current state of knowledgein the field and its relevance for clinicians. Even though thescope of the article is limited to research on language andtheory of mind, it is not feasible to do a thorough review, as theliterature on theory of mind is vast and has far-reachingimplications for many aspects of social, cognitive, and lin-guistic development. The publications cited here will providea point of entry into the literature for the reader who wishes tolearn more. First, a definition of theory of mind is provided,

along with a brief overview of its developmental course, and

a description of tasks that are often used to assess theory of mind. Next is a review of ways in which theory of mind andlanguage are thought to influence one another in development,and an overview of the role of language in theory of mindassessment. Evidence regarding theory of mind and languagein atypical populations is described. In the final section, someclinical implications of the literature are suggested and illus-trated with a hypothetical case study.

What Is a Theory of Mind?

Theory of mind refers to an understanding of mentalstates—such as belief, desire, and knowledge—that enables

us to explain and predict others’ behavior. Consider thefollowing scenario. I make some cookies and bring them towork. I show my coworker, George, that I am putting thecookies in my desk drawer and invite him to help himself anytime. A little later, I notice a couple of ants crawling near my desk and decide to move the cookies to a high shelf. Thatafternoon, on my way to the mailroom, I see George headedtoward my office. I say, ‘‘They’re on the shelf.’’ I haveeffortlessly computed that George wants a cookie but becausehe incorrectly believes the cookies are in the drawer, he willfail to find them; if I tell him where ‘‘they’’ are, I will changehis belief, enabling him to fulfill his desire for a cookie. I canexplain and predict George’s behavior in terms of his mental

states, because I have a theory of mind. This sort of reasoning

American Journal of Speech-Language Pathology Vol. 15 142–154 May 2006 A American Speech-Language-Hearing Association

1058-0360/06/1502-0142

142

Tutorial

Developmental Relationships Between

Language and Theory of Mind

Carol A. MillerThe Pennsylvania State University,University Park

Purpose: This tutorial is intended to informreaders about the development of theory of mind(understanding of mental states) and to discussrelationships between theory of mind and lan-guage development.Method: A narrative review of selected literatureon language and theory of mind is presented.Theory of mind is defined, and commonly usedmeasures of theory of mind are described.Developmental relationships between languageand theory of mind in typical and atypical popula-tions are discussed. Literature-based suggestionsfor clinical assessment and intervention areprovided, using a hypothetical case study.

Conclusions: The article serves as an introduc-tion to current research about language andtheory of mind, and emphasizes their interde-pendence in development. Implications of therelationships between theory of mind and lan-guage development for language assessmentand intervention are discussed, and an argumentis made that taking theoryof mind into account willhelp clinicians enhance children’s communicationand language development.

Key Words: cognition, normal languagedevelopment, language disorders,autism spectrum disorders



is complex when made explicit, yet we do it all the time, withlittle or no conscious reflection. Developmental evidence,however, suggests that such thinking is not automatic for young children, who must develop a number of skills in order to reach the adult level of competence in understanding of mental states.

Precursors of Theory of Mind

Theory of mind is a broad construct that is reflected inmany kinds of knowledge and skills. It is not all-or-none. Likelanguage, theory of mind develops over time, building fromfoundational, precursor skills to a sophisticated understandingof how mental states and behavior interact. Precursors of theory of mind include joint attention, appreciation of inten-tionality, recognition that different people have different per-spectives, use of mental state words, and pretend play. Anapproximate timeline is shown in Table 1.

Joint attention, defined by Morales et al. (2000) as ‘‘thecapacity of an infant to coordinate her attention with a socialpartner vis-a-vis an object or event’’ (p. 283), is consideredto be an early-developing component of theory of mind(Tomasello, 1995). By 9 months, infants begin to share withanother person the experience of attending to something(Carpenter, Nagell, & Tomasello, 1998) and, according toTomasello (1995), come to understand that others intend for the infant to pay attention to a specific aspect of an object or event. Any given object or event has an infinite number of characteristics that can be commented on. A communicativeact demonstrates the communicator’s intention to single outa very small subset of these characteristics for the listener tofocus on. The understanding of this intentional property of communication comes to fruition in the use of words, which

are used to pick out particular aspects of an object or event,such as red, or ball, or fast (Tomasello, 1995). It is logical, asarguedby Tomasello, that joint attention, because of its role inappreciation of intentional communication, is a precursor of theory of mind; however, as will be discussed below, the con-nection currently has more theoretical than empirical support.

In the preschool years, children begin to learn that differ-ent people may have different interpretations of the sameobject, depending on perspective (Flavell, Everett, Croft, &Flavell, 1981). For example, when two people are facing oneanother across a table, a picture that is flat on the table willappear right-side up to one and upside down to the other

(Flavell et al., 1981). Children also learn that not everyonelikes or wants the same things (Flavell, Flavell, Green, &Moses, 1990; Repacholi & Gopnik, 1997). The understandingthat perspectives can differ is essential for successful com-munication; a communicator has an intention to take a certainperspective on an object, but not necessarily the same per-

spective as the listener.During the 3rd year of life, young children begin to talk

about mental states. Their early use of mental state termssuggests less than full comprehension of the terms’ mentalimplications, as children use phrases such as ‘‘I think’’ asan equivalent of ‘‘maybe,’’ and ‘‘Know what?’’ to initiate aconversation (Shatz, Wellman, & Silber, 1983). The use of mental state terms gradually becomes more truly mental-istic, unambiguously referring to the thoughts, beliefs, andfeelings of oneself and others (Bartsch & Wellman, 1995;Shatz et al., 1983).

Children begin to engage in pretendplay as early as the 2ndyear of life, and it has been argued that pretend play is an

important component of theory of mind development becauseit requires one to ‘‘decouple’’ a representation of reality (e.g., awooden block is a car) from reality itself (Leslie, 1987, 1994;see below for further discussion of representations). Fewstudies have examined longitudinal relationships betweenpretend play and theory of mind development. Youngbladeand Dunn (1995) found that acting out roles in pretend playat age 2;9 (years;months) predicted false belief performanceat 3;4, even when language (measured by mean length of utterance [MLU]) was controlled. Astington and Jenkins(2000), however, found that none of their measures of pretendplay (total amount of pretending, joint planning of pretend,and explicit assignment of roles) predicted theory of mindperformance, although theory of mind predicted later jointplanning and role assignment. Astington and Jenkins’s par-ticipants ranged in age from 2;10 to 3;9 at the beginning of the study and were tested three times over a 7-month period.Given the conflicting results of these studies, further researchis needed to explicate the relationship between pretend playand theory of mind in development.

Assessing Theory of Mind

Children enter the preschool years with many skills to forma foundationfor theory of mind. Between 3 and 5 years of age,further important developmental changes in theory of mind

TABLE 1. Approximate developmental timeline of some aspects of theory of mind, with illustrative references.

Age Aspects of theory of mind

6–12 months & Joint attention, including gaze and point following, and alternation of gaze between person and object(Bruinsma et al., 2004; Carpenter et al., 1998)

& First words (Tomasello, 1995)13–24 months & Recognize intentionality in others as demonstrated in word use (Tomasello, 1995)

& Recognize that others have desires different from one’s own (Repacholi & Gopnik, 1997)& Early pretend play (Leslie, 1987)

30–36 months & Begin to use mental state terms with truly mentalistic functions (Bartsch & Wellman, 1995)& Increasingly sophisticated pretend play (Youngblade & Dunn, 1995)

37–48 months & Increasing ability to understand how things look from another’s perspective (Flavell et al., 1981)& Begin to understand sentence complements (de Villiers & Pyers, 2002)

49–60 months & Consistently pass false belief and appearance-reality tasks (Wellman et al., 2001)

Miller: Language and Theory of Mind 143



take place (see Table 1). Children begin to understand thatour minds do not simply mirror reality, but that we activelyconstruct representations of reality. These representationsmay change within an individual as new information isreceived, and they may differ across individuals, dependingon the information each person has access to (Astington,

1993; Bartsch & Wellman, 1995). Importantly, mental rep-resentations sometimes fail to represent reality accurately.In the example of George and the cookies, George has con-structed a representation of reality based on seeing me put thecookies in the drawer. This representation becomes false, or inaccurate with regard to reality, when I move the cookies.George does not have access to this information. By tellingGeorge the current location of the cookies, I give him infor-mation that enables him to revise his representation of reality.

Several methods have been used to assess the ability todistinguish between representation and reality. One com-monly used method is the appearance-reality task (Flavell,1992). This task uses deceptive objects, that is, objects whose

appearance belies their true nature. For example, a child isshown a sponge painted to look like a rock. Before beingallowed to touch it, the child is asked what it is, and she saysthat it is a rock. Then she is encouraged to touch and manip-ulate the object, and when asked, she says that it is a sponge.Two test questions are then asked: ‘‘What does it look like?’’and ‘‘What is it, really and truly?’’ Three-year-olds usuallygive the same answer to both questions, either saying that itlooks like and is a rock, or that it looks like and is a sponge.Four-year-olds are usually able to answer the two questionsdifferently (and correctly), saying that the object looks likea rock but is a sponge.

Researchers have also focused on preschoolers’ growingunderstanding of false belief—in other words, the understand-ing that a person may have a belief (or representation) thatis different from reality and will act in accordance with thatbelief. Two types of tasks often used to assess false belief understanding at this age are change of location and unex-pected contents. In a typical change of location task, a dollnamed Sally places a ball in a basket, then leaves. A seconddoll moves the ball to a box in Sally’s absence. The childis asked where Sally thinks the object is. Three-year-oldsusually answer incorrectly, saying that Sally thinks the ballis in the box. Four-year-olds usually answer correctly, sayingthat Sally thinks the ball is in the basket. In orderto succeed onthe task, a child must recognize that Sally’s representationof the world (ball in basket) hasbecome false, andthat she will

act in accordance with that representation, or belief, rather than in accordance with current reality, or what the childknows to be true. Descriptions of change of location tasks canbe found in Astington and Jenkins (1999), Miller (2001), andWimmer and Perner (1983), to name just a few.

In a typical unexpected contents task, the child is shown acrayon box. The child is asked what she thinks is in the box,and she answers, ‘‘Crayons.’’ The experimenter opens the boxand shows the child that it actually contains paper clips. Theexperimenter then closes the box and asks the child, ‘‘Whatdid you think was in the box before we opened it?’’ Three-year-olds usually say that they thought there were paper clipsin the box. The experimenter also asks, naming one of the

child’s friends, ‘‘If your friend came in here and saw this box,

what would he think is in it?’’ Again, 3-year-olds usually saythat the friend will think the box contains paper clips. Four-year-olds usually answer both the questions about their ownprior belief and their friend’s belief correctly. Examples of unexpected contents tasks can found in Farrar and Maag(2002) and Gopnik and Astington (1988).

The basic forms of the change of location task and theunexpected contents task have been varied in many ways,in order to determine what parameters affect children’s per-formance. In change of location, control questions are usuallyasked to make sure the child remembers where the ball(or other object) was placed initially, and where it ended up.Some parameters that are varied in both tasks include thewording of the questions, whether the transformation inlocation or contents was presented as a ‘‘trick,’’ and in thechange of location task, whether the protagonist is a realperson as opposed to a doll or puppet. According to Wellman,Cross, and Watson (2001), such manipulations have little im-pact on the age at which children succeed on false belief tasks.

The finding that children become able to pass false belief tasks consistently at around 4 to 5 years of age is a robustone (see Table 1).

As children grow older, more complex false belief taskscan be used to assesstheory of mind. The traditional change of location and unexpected contents tasks measure understand-ing of first-order false belief—that is, understanding oneperson’s belief. Second-order false belief involves a person’sbelief about someone else’s belief (Perner & Wimmer, 1985).For example, suppose that before work, my husband and Iagree that in the afternoon, we will meet at Pine Park tosee a rugby game in which our friend Leon is playing. After Iget to work, Leon tells me that the game has been moved toOak Park. I have meetings the rest of the day and do not havea chance to call or e-mail my husband about the change.Meanwhile, unknown to me, Leon bumps into my husbandat lunch and tells him of the location change. At the end of the day, I try to call my husband, but it is too late. He hasalready left home, and I have our cell phone. Where will Igo to meet him—Oak Park or Pine Park? The reader with amature theory of mind will easily solve this second-order falsebelief problem: I will go to Pine Park, falsely believing thatmy husband has a false belief about the location of the rugbygame. Stories such as this, sometimes accompanied by pic-tures or an enactment with toy figures, form the basis for assessing second-order false belief understanding.

Language and Theory of Mind in Development

Throughout the first few years of life, the development of language and theory of mind are interwovenin complex ways.Infants engage in joint attention and demonstrate appreciationof others’ intentions within the context of communicativeacts. Toddlers begin to use mental state terms in increasinglymore mentalistic ways and engage in pretend play. Youngchildren begin to understand that different people have dif-ferent access to information and different desires. They listento and participate in conversations in which people predictand explain behavior in terms of desires, beliefs, and feelings.This section of the tutorial describes some ways that language

and theory of mind interact in development.

144 American Journal of Speech-Language Pathology Vol. 15 142–154 May 2006



Both language and theory of mind undergo rapid anddramatic developmental changes during the first 5 years of life, but so do other domains; therefore, the temporal associ-ation is not enough to support the hypothesis that languageand theory of mind are related. There are reasons, however,to expect development in language and theory of mind to

be related. As noted at the beginning of the article, successfulcommunication requires an appreciation of the mental statesof the interlocutor. Theory of mind is necessary for commu-nication through language, but language may in turn offer away to learn about theory of mind.

One reason language is important for theory of minddevelopment is that mental states are unobservable. We maybe able to learn the meaning of a word like run by observingwhat happens when the word is said, but we certainly cannotlearn the meaning of a word like think by observing whathappens when it is said (Gleitman, 1990). Thinking, knowing,liking, wanting, and so on are internal activities or states thatdo not have reliable behavioral correlates. Language, there-

fore, is a crucial source of information about what mental stateterms mean (Gleitman, 1990). The information availableabout mental state terms comes from their roles in the gram-matical and semantic systems of a language (e.g., the typesof sentences in which they occur) and from their roles in thepragmatics of a language (e.g., when parents say things like‘‘Your brother is mad because he thought it was his turn’’).This section describes some of the empirical evidence linkinglanguage and theory of mind development. It should be notedthat most of the data come from English-speaking childrenand their families. Cross-linguistic and cross-cultural researchin language and theory of mind is limited, but growing (e.g.,Lohmann & Tomasello, 2003; Perner, Sprung, Zauner, &Haider, 2003; Shatz, Diesendruck, Martinez-Beck, & Akar,2003; Tardif, Wellman, & Cheung, 2004).

Relationships Between Language Exposure

and Theory of Mind

A number of studies have suggested that children’s theoryof mind development is influenced by their exposure to talkabout mental states. Ruffman, Slade, and Crowe (2002) foundthat mothers’ talk about mental states predicted children’slater theory of mind performance, as did the children’s lan-guage ability. Children’s earlier theory of mind performance,however, did not predict later mental state talk by mothers,suggesting a causal role for mothers’ talk about mental

states in their children’s theory of mind development. It isnot only mothers who may play a role in theory of minddevelopment. Dunn, Brown, Slomkowski, Tesla, andYoungblade (1991) found that theory of mind understandingat 40 months was correlated with engagement in family talkabout feelings and causality, and cooperative interaction witha sibling, at 33 months. There is evidence that children withsiblings are advantaged in theory of mind development (e.g.,Jenkins & Astington, 1996; Lewis, Freeman, Kyriakidou,Maridaki-Kassotaki, & Berridge, 1996; Perner, Ruffman, &Leekam, 1994; Peterson, 2000), presumably because of theopportunities for discourse and experiences related to others’thoughts and feelings that siblings provide.

Further evidence that is consistent with a role for familytalk in the development of theory of mind comes from deaf children of hearing parents who are late learners of sign lan-guage. Woolfe, Want, and Siegal (2002) found that latesigners (wholearned sign languagein school) performed morepoorly on a theory of mind task than native signers. The

authors argued that late-signing deaf children do not haveas many opportunities for family discourse as hearing childrenof hearing parents, or deaf children of deaf parents, and hencefewer opportunities to learn about mental states through con-versation. Overall, then, evidence from typical and atypicaldevelopment suggests that opportunities to listen to and en-gage in conversations about mental states contribute to thedevelopment of theory of mind.

Joint Attention, Language Development,

and Theory of Mind

In recent years, researchers and theorists have increasingly

emphasized the role of joint attention in the development of both theory of mind and language (Moore & Dunham, 1995;Sigman & Ruskin, 1999). The ability to respond to adult bidsfor joint attention has been found to be associated with vocab-ulary development in children with typical development (e.g.,Carpenter et al., 1998; Morales et al., 2000; Morales, Mundy,& Rojas, 1998), as well as in children with autism spectrumdisorder, Down syndrome, and developmental delay (Sigman& Ruskin, 1999). Rollins and Snow (1998) found that theability to engage in joint attention was associated with gram-matical development in both children with typical develop-ment and childrenwith autism. Childrenwith autism spectrumdisorder exhibit early deficits in joint attention, and this seemsto be a crucial aspect of their communicative deficit (e.g.,

Bruinsma, Koegel, & Koegel, 2004; Rollins, Wambacq,Dowell, Mathews, & Reese, 1998; Sigman & Kasari, 1995;Sigman & Ruskin, 1999).

For children with typical development learning their firstwords, it is easiest to learn an object label when an adultprovides a name for an object that is already the child’s focusof attention (Tomasello & Farrar, 1986). By 18 months, how-ever, an infant with typical development can monitor whatan adult is attending to when the adult utters a new word, andform hypotheses about the word’s meaning accordingly— even if the adult’s focus of attention is different from thechild’s (Baldwin, 1995).

Charman et al. (2000) noted that, although the relationships

between joint attention and language and between languageand theory of mind have been taken as indirect evidence that

joint attention is related to later theory of mind development,direct evidence of such a relationship was lacking. In a sampleof 13 children with typical development, Charman et al. foundthat a measure of joint attention (shifting gaze between anadult and an interesting toy) at 20 months was positively cor-related with theory of mind performance at 44 months. Thus,there is a substantial body of evidence supporting joint atten-tion as a precursor of language development, and a logicalconnection between joint attention and theory of mind vialanguage; however, further investigation of a direct relation-ship between joint attention and theory of mind is needed.




Mental State Terms

Since the early days of research on theory of mind, investi-gators have examined children’s use of language for evidenceof mental state understanding. Shatz et al. (1983) studied theemerging use of mental state terms by toddlers. They foundthat cognition terms such as

know, think, mean, forget,and

guess were first used by children between 2;4 and 2;8; how-ever, examination of the contexts in which these words wereused suggested that they did not have true mental state func-tions, but occurred in the routinized phrase ‘‘I don’t know’’ or were used to manage discourse (e.g., ‘‘Know what?’’ to ini-tiate conversation or take a conversational turn).

Terms expressing desire may be among the first to be usedwith a truly mentalistic function. Bretherton and Beeghly(1982) found that by 2;4, more than half of the children in their sample applied the desire terms want and need to both self andothers, and Bartsch and Wellman (1995) observed genuinereference to desire soon after the second birthday. By compar-ison, Bartsch and Wellman (1995) found that think and know

were not used as true mental state terms before 2;7. True mentalstate functions for cognition words were not observed byShatz et al. (1983) before 2;6. Both Shatz et al. and Bartschand Wellman relied on analysis of surrounding context toidentify genuinely mentalistic use of mental state terms. Themost convincing mentalistic uses are contrastive uses. Theseuses contrast someone’s mental state with reality, contrastone’s mental state with one’s own prior mental state, or con-trast someone’s mental state with someone else’s mental state(Bartsch & Wellman, 1995; Shatz et al., 1983). An examplefrom Bartsch and Wellman (1995, p. 46), produced by a childat age 3;8, is ‘‘I thought I could rip the papers off, ‘cept itdoesn’t have any paper.’’ Evidence that mastery of mental

state language is related to theory of mind comes from an ex-perimental study by Moore, Pure, and Furrow (1990), whichshowed that 4-year-olds’ understanding of the relative cer-tainty implied by the verbs think and know was related totheir false belief performance.

General Language Ability

In a seminal study following children from age 3;4 to3;11, Astington and Jenkins (1999) measured theory of mindand administered the Test of Early Language Development(Hresko, Reid, & Hammill, 1981). Astington and Jenkinsfound that scores for syntax, but not for semantics, predictedlater performance on theory of mind measures. The reversewas not true; earlier theory of mind performance did notpredict later language scores. Expressive and receptive scoreswere combined in these analyses, but when expressive andreceptive syntax were examined separately, their predictivevalue was similar.

Further studies have confirmed that language predictslater theory of mind, but they have cast doubt on the uniqueimportance of syntax. Farrar and Maag (2002) obtained parentreport measures of vocabulary size and expressive grammat-ical complexity, as well as MLU, when children were 27months old, then measured theory of mind at age 4 years,using appearance-reality, unexpected contents, and changeof location tasks. They found that vocabulary and MLU

predicted theory of mind performance, but grammatical com-plexity was not a predictor when they controlled for vocab-ulary. Ruffman, Slade, Rowlandson, Rumsey, and Garnham(2003) compared measures of syntax and semanticsat age 3 aspredictors of theory of mind performance at ages 3O, 4, and5O. Ruffman et al. found that semantic ability accounted for

unique variance in understanding of belief, but syntactic abil-ity did not account for additional variance.

It seems clear that language and theory of mind perfor-mance are related, and that language is a better predictor of theory of mind than the reverse. It is less clear what aspects of language development predict theory of mind development,as some studies show a greater influence of semantics, andothers show a greater influence of syntax. Some researchershave focused on a particular syntactic construction as a keyinfluence on theory of mind.

Sentence Complements and False Belief

It has been suggested that a specific aspect of syntax, rather than syntactic ability in general, is crucial for false belief understanding. The hypothesis proposed by J. de Villiers andcolleagues (de Villiers & de Villiers, 2000; de Villiers &Pyers, 2002) posits that understanding of sentence comple-ment structures is necessary for children to be able to under-stand false belief. An example of a sentence complementstructure is ‘‘Lucy thinks the moon is made of green cheese.’’The sentence as a whole may be true even though thesentence embedded within it (‘‘the moon is made of greencheese’’) is false. The truth value of the entire sentence isindependent of the truth value of the embedded sentence.Only certain verbs allow this type of construction, includingmental verbs such as think, believe, and guess, and commu-

nication verbs such as say and tell. Until children under-stand that ‘‘Lucy thinks the moon is made of green cheese’’can be true while ‘‘the moon is made of green cheese’’ is false,they have no way to represent false beliefs and thereforemust fail false belief tasks. In a longitudinal study, de Villiersand Pyers (2002) found that the ability to understand sen-tence complements predicted false belief performance 3 to4 months later.

Both mental verbs and communication verbs were exam-ined by de Villiers and Pyers (2002), and that study suggestedthat communication verbs serve as a bootstrap for sentencecomplement mastery. The truth value of the embedded com-plement in sentences with communication verbs can often

be verified. If a child hears ‘‘Mom said we’re having chickenfor dinner’’ but then is served hamburgers instead, she hasverifiable evidence that the truth value of the embeddedsentence is independent of the truth value of the entire sen-tence. This understanding will generalize to mental verbs,which use the same structure.

Language Intervention and False Belief

One way to test what kinds of language experiences lead tofalse belief understanding is through intervention. Hale andTager-Flusberg (2003) found that providing 3- to 4-year-oldchildren with training on sentence complements improvedfalse belief performance as much as direct training on false




belieftasks. Training on false belief, however, did not result inimproved understanding of sentence complements. Lohmannand Tomasello (2003) also found that sentence complementtraining resulted in better false belief performance. Further-more, false belief improved following training with discourseabout deceptive objects that did not include sentence com-

plements. In Lohmann and Tomasello’s study, the most im-provement was found when training included both sentencecomplements (with either mental or communication verbs)and experience with deceptive objects. The results of thesetwo studies suggest that sentence complement comprehensiontraining is sufficient, but not necessary, for improved false be-lief understanding in preschoolers with typical development.

Interventions emphasizing discourse about mental statesand false beliefs have had some success. Appleton and Reddy(1996) had children view videos in which someone has a falsebelief and discuss them with an experimenter. The children inthe training group improved more on false belief tasks than acontrol group who had storybooks read to them by the exper-

imenter. Guajardo and Watson (2002) used book reading astraining, where the experimenter read each child books in-cluding mental state references, discussed the mental stateaspects with the child, and encouraged acting out of the story.Guajardo and Watson found more false belief improvementfor the training group than a no-training control group. Peskinand Astington (2004) also used book reading as training,but they edited the books to make the mental state contenteither explicit or implicit. Although the group that wasexposed to the explicit books improved more on the use of mental state terms, the group that was exposed to the implicitbooks improved more on a task requiring explanation of acharacter’s false belief. Both groups improved on predictionof false belief (change of location and unexpected contentstasks), but neither outperformed the other.

Summary

The majority of the evidence pertaining to relationshipsbetween language development and theory of mind is corre-lational in nature; therefore, it is premature to say that certainlanguage and communication experiences or milestones arenecessary for theory of mind development. Nonetheless, apicture is emerging of complex interdependencies betweenlanguage and theory of mind, beginning in infancy with jointattention and appreciation of intentionality, and continuing astoddlers begin to use mental state terms in increasingly men-talistic ways. Young children listen to and participate in con-versations in which people predict and explain behavior interms of desires, beliefs, and feelings. Language skills growand support a developing theory of mind, while at the sametime, the increasingly sophisticated theory of mind makes itpossible to engage in meaningful communication. Movingbeyond correlational analyses, evidence from interventionstudies suggests that certain language experiences result inimproved false belief performance.

Language in Theory of Mind Assessment

No doubt it will have occurred to the reader that consider-able language ability is required simply to participate in

standard theory of mind tasks. This potential artifact has notgone unnoticed by researchers, and many experimental ma-nipulations have been used to reduce or remove the languagedemands of false belief tasks. These language demands maycome from the syntactic complexity of the test question, thepragmatics of the test question, or the overall language com-

prehension required to follow the experimenter’s explanation.In false belief tasks, the crucial data point is the children’s

response to a test question. This question is usually a sentencecomplement construction, such as ‘‘Where does Sally thinkthe ball is?’’ It seems possible that the complexity of thesentence complement could tax the resources of the children,leading them to answer in a way that does not reflect their actual understanding of false belief. In response to this line of reasoning, a common change to the procedure is to manipulatethe phrasing of the test question, asking (in change of locationtasks) ‘‘Where will Sally look for the ball?’’ In their meta-analysis, Wellman et al. (2001) found that this change did notaffect the age at which children succeed on the tasks.

The ‘‘look’’ question, however, raises potential problemsof its own with regard to discourse pragmatics. Siegal andBeattie (1991) focused on this issue, using false belief stories(enacted using toy figures in a dollhouse) similar to change of location tasks, such as ‘‘Jane wants to find her kitten. Janethinks her kitten is in the kitchen. Jane’s kitten is really in thebathroom. Where will Jane look for her kitten?’’ (p. 3). Theyreasoned that the question ‘‘Where will Jane look for her kitten?’’ could be interpreted as ‘‘Where will Jane have to lookfor the kitten in order to find it?’’ leading to the error of choos-ing the kitten’s current location. Therefore, the researchersasked, ‘‘Where will Jane look first for the kitten?’’ and foundthat children performed more accurately in this condition.

Rewording the question to reduce ambiguity has also beentried in unexpected contents tasks. In such tasks, children areshown the true contents of a deceptive container and thenasked what they had thought was in the container. Lewis andOsborne (1990) hypothesized that children might not interpretthe question as being about their earlier belief state. If this wastrue, performance should be improved by adding temporalspecificity to the test question. After eliciting the child’s guessas to the contents of a candy container and then showing thechild the actual contents (a pencil), Lewis and Osborne askedone of three questions: ‘‘What did you think was in the box?’’or ‘‘What did you think was in the box when the top was stillon it?’’ or ‘‘What did you think was in the box before I took thetop off?’’ Three-year-olds were more accurate when the

question included the before phrase.Lewis and Osborne’s (1990) results suggest that increasing

syntactic complexity does not negatively affect performanceof children with typical development if it clarifies the testquestion. Two studies of children with specific languageimpairment (Miller, 2001, 2004) have tested the hypothesisthat false belief competence might be masked in performanceby the difficulty of understanding the test question. For thesechildren, asking ‘‘Where will Sally look for the ball?’’ resultedin better performance relative to age-matched controls than‘‘Where does Sally think the ball is?’’ Younger controlsmatched for language comprehension did not benefit from thechange in the test question.




Some language demands of false belief tasks are not easyto categorize as syntactic or pragmatic. False belief tasks ofteninclude a narration by the experimenter, which may be fairlyelaborate; see Wimmer and Perner’s (1983) change of loca-tion task involving Maxi and his chocolate for an example.Some researchers have attempted to minimize the language

demands of false belief tasks by creating less verbal or non-verbal versions. De Villiers and de Villiers (2000) developed atask that used pictures to depict a change of location scenarioand required the child to choose a surprised facial expressionfor the protagonist who, unaware that a desired object hadbeen moved, looked for it in its original location. Typicallydeveloping 3- and 4-year-old children performed no better onthis task than on a traditional, more verbal task. The task wasalso given to a group of orally educated deaf children withlanguage delay, ranging in age from 5 to 10 years. Thesechildren performed worse on the less verbal task than a com-parable verbal task. Reducing the language demands of thetask did not improve performance.

Call and Tomasello (1999) developed a nonverbal falsebelief task that could be used with chimpanzees and, withminimal adaptation, preschoolers. In the task, an adult exper-imenter gave visual cues to the location of a sticker. Thechildren were unaware of the sticker’s true location, but infor-mation was available that should tell them that the adult had afalse belief about the location. Therefore, if they understoodfalse belief, the children should look elsewhere for the sticker.Performance on the nonverbal task was comparable to averbal version.

In summary, simplifying the language used in false belief tasks sometimes results in improved performance, but a com-pletely nonverbal task is no easier than a verbal one for chil-dren with typical development (Astington & Baird, 2005;Astington & Jenkins, 1999). The role of language demands infalse belief may be more complex for children with languagedisorders (Miller, 2004). Accordingly, it is important to becareful how we assess theory of mind. The performance of some children can be enhanced by manipulating the languageused in theory of mind tasks, although it seems clear thatthe robustness of general developmental trends in theory of mind is not merely an artifact of how the tasks are presented.

Theory of Mind and Communication Disorders

Some clinical populations with communication disordershave already been mentioned in the review of research on

language and theory of mind. Children with autism spectrumdisorder are the most intensely studied clinical population inthe theory of mind literature, and other populations are oftenincluded as comparison groups, including children with mentalretardation (e.g., Fisher, Happe, & Dunn, 2005; Happe, 1995;Tager-Flusberg, 2000) and children with specific languageimpairment (e.g., Gillott, Furniss, & Walter, 2004; Perner,Frith, Leslie, & Leekam, 1989; Tager-Flusberg & Joseph,2001; Ziatas, Durkin, & Pratt, 1998). The results of such stud-ies generally show that children with developmental disordersother than autism spectrum disorder perform better on theoryof mind tasks, but not necessarily at age-appropriate levels.

A few studies have examined children with language

deficits directly, to address questions about the relationships

between language and theory of mind. Peterson and Siegal(1995) andde Villiers and de Villiers (2000) have studieddeaf children with and without early exposure to sign language.Those without early exposure have language deficits andperform poorly on false belief tasks. Studies that have focusedon theory of mind in children with specific language impair-

ment (Bishop, 1997; Farmer, 2000; Miller, 2001, 2004; Shields,Varley, Broks, & Simpson, 1996; van der Lely, Hennessey, &Battell, 1999) have found that those whose language prob-lems did not include pragmatic deficits performed at or near age levels on theory of mind. In all of these studies exceptthose of Miller (2001, 2004), however, the children were wellpast preschool age.

Tager-Flusberg (2000) argued that the false belief perfor-mance of children with autism spectrum disorder is relatedto language ability in ways that may be unique. Grammat-ical ability may be more important for theory of mind inchildren with autism spectrum disorder (see also Fisher et al.,2005). As suggested earlier, there is some evidence that

sentence complements play a key role in the development of false belief abilities. Tager-Flusberg suggested that sentencecomplements may be even more important for children withautism spectrum disorder, who build on communication verbsand the complement structures in which they appear as a wayto achieve success on false belief tasks. Because they lackinsight into mental states, children with autism spectrum dis-order may use sentence complements with communicationverbs to bootstrap an understanding of similar structures withmental state verbs. This hypothesis is consistent with deVilliers and Pyers’s (2002) finding that in children withtypical development, understanding of sentence complementstructures with communication verbs was a better predictor of false belief performance than sentence complements withmental verbs. Children with autism spectrum disorder, how-ever, may rely more heavily on communication verbs.Johnston, Miller, and Tallal (2001) found evidence that theuse of communication verbs may be a strength for childrenwith specific language impairment also. Analyzing sponta-neous language samples, Johnston et al. found that childrenwith specific language impairment produced no more cog-nitive state predicates than younger, language-matched con-trols, but did produce more predicates about communicationevents (e.g., say, tell, point ).

Clinical Implications

Theory of mind begins early, as children share jointattention with adults and begin to talk about mental states. Astoddlers become preschoolers, they become able to under-stand that people have mental representations of the worldbased on experience, that these representations may some-times be false, and that behavior is driven by these represen-tations. In concert with these changes in theory of mind,children are also becoming more sophisticated languageusers, and indeed much of what children know about therelations of mental states to behavior may derive from their exposure to, and engagement in, conversations.

What does all this mean to clinicians as they work withchildren? First and foremost, clinicians must be aware that

theory of mind helps form the foundation for communication




that is not merely instrumental—a way of fulfilling needs andwants—but involves a rewarding exchange of ideas, feelings,and information. The relationship between language andtheory of mind is a two-way street, however. Conversationsprovide a crucial context for learning about the mind, andif children’s communication disorders limit their ability to

engage in, and benefit from, such conversations, theory of mind development is at risk, and with it, further communica-tive development. In addition, there is evidence that complexsyntax, especially the sentence complement structure, pro-vides an important tool for thinking about false belief. Lan-guage and theory of mind build on one another; therefore,theory of mind can be a concern not only for children withautism spectrum disorder, who are known to be at risk for bothlanguage and theory of mind deficits (Baron-Cohen, Tager-Flusberg, & Cohen, 2000), but also for children with specificlanguage impairment, mental retardation, or general develop-mental delay.

Although few standardized instruments exist for the eval-

uation of theory of mind and related skills, the research liter-ature offers some directions for assessment and intervention.In most cases, these directions are founded on inferencesfrom correlational studies, rather than direct evidence fromintervention experiments, although some intervention data areavailable. This section suggests ways clinicians may assesssome precursor and component skills of theory of mind, andaddresses possibilities for intervention. It is not unreasonableto ask why speech-language pathologists should take the timeto assess theory of mind at all. One answer is that speech-language pathologists treat the whole child. Just as they con-cern themselves with cognitive skills, such as memory, thatare separate from language but closely related to it, theyshould consider the development of theory of mind, which isdistinct from language but at the same time deeply intercon-nected with it. As will be demonstrated in the hypotheticalcase study below, by understanding a child’s theory of mindabilities, the clinician can select appropriate language goalsthat are also likely to promote theory of mind development.

Joint Attention

For children who are very young developmentally or chro-nologically, observation of behavior is important to determinewhether the child is able to follow an adult’s attentional lead,make a bid for the adult’s attention, or engage in joint atten-tional episodes by alternating gaze between an object and an

adult. Gaze following can be assessed by sitting opposite thechild, making eye contact, and then looking toward a targetlocated to the side, while looking interested and/or vocalizing(e.g., gasping or calling the child’s name). Observe whether the child follows a clinician’s or a parent’s gaze and how longit takes. Carpenter et al. (1998) and Morales et al. (2000) pro-vide excellent descriptions of methods for assessing jointattention.

Published instruments are available that include assess-ment of joint attention. The Communication and SymbolicBehavior Scales and Communication and Symbolic Behavior Scales Developmental Profile (Wetherby & Prizant, 1993,2002) are norm-referenced, using standardized procedures toobserve and elicit communicative behaviors. The Early Social

Communication Scales (Mundy et al., 2003; Seibert, Hogan,& Mundy, 1982) also provide a standardized method for observation of joint attention. A few studies have investi-gated intervention for joint attention with children with autismspectrum disorder, with some success. For example, Rollinset al. (1998) described a naturalistic approach in a case study,

whereas Whalen and Schreibman (2003) used a behavior modification approach.

Mental State Terms

Spontaneous language samples can be used to determinewhether a child is talking about mental states. Many children’spicture books include reference to mental states (Cassidyet al., 1998; Dyer, Shatz, & Wellman, 2000). Cassidy et al.examined 317 children’s books and found that 78% containedreferences to internal states such as thinking, knowing, feel-ing, wanting, and so on; 34% contained references to falsebelief. Book reading, then, can provide ample opportunities

for talk about mental states.Lack of a vocabulary to talk about mental states may limitchildren’s theory of mind development and communication.Just as shared reading of children’s literature may provide away to assess the use of mental state terms, it may also be agood way to increase use of mental state terms. Peskin andAstington (2004) edited the text of six picture books so that thetheory of mind elements were either explicit or implicit. Four-year-olds were exposed to either the explicit or implicit theoryof mind books approximately 72 times on average over aperiod of 4 weeks. Following intervention, the group exposedto the explicit theory of mind books produced more mentalstate terms in a story-telling task.

Earlier, evidence was reviewed showing that the everyday

talk about mental states that takes place in families is asso-ciated with theory of mind development. Intervention studiesusing such talk as a treatment would be needed to determinewhether the association is causal in nature. No such studieshave been done; however, it seems likely that conversationsabout mental states and their relation to behavior would beappropriate as a component of intervention. Such conversa-tion might be easily incorporated in pretend play, or in theeveryday routine of a preschool setting, where there is livelyinteraction among peers as well as between adults andchildren.

False Belief and Appearance-Reality Tasks

This tutorial has described several commonly used theoryof mind tasks: appearance-reality, change in location, andunexpected contents, as well as (for older children) second-order false belief. The basic tasks are summarized, along withsome common variations where appropriate, in Table 2.Sources that describe the tasks in more detail are also pro-vided. Use of these tasks should enable clinicians to better understand their clients’ theory of mind abilities. As the tradi-tional forms of these measures have significant languagedemands, clinicians should be cautious in assessing childrenwith atypical language development. Although nonverbalversions of false belief tasks do not seem to confer an advan-tage (Astington & Jenkins, 1999; Call & Tomasello, 1999;




de Villiers & de Villiers, 2000; Miller, 2004), there is someevidence that children with specific language impairment aresensitive to the phrasing of the test question in a change of location task (Miller, 2001, 2004).

It was noted earlier in this article that language appearsto be a better predictor of theory of mind than the reverse.Therefore, speech-language pathologists may want to imple-ment intervention activities that serve the dual purpose of building language competence (especially in the areas of sentence complement structures, mental state and commu-nication terms, and understanding and telling stories) andenhancing theory of mind abilities. A handful of trainingstudies with typically developing 3- and 4-year-old children

have shown that different types of language experiences

resulted in improvement on false belief tasks. Hale and Tager-Flusberg (2003) and Lohmann and Tomasello (2003) dem-onstrated that training on sentence complement structures ledto better false belief performance. Lohmann and Tomaselloalsoshowed thatusing sentence complements to talk about thenature of deceptive objects(e.g., a pen that looks like a flower)led to improvement beyond the effect of sentence comple-ments alone. There was no difference in the results whether mental or communication verbs were used in the sentencecomplements. There is also empirical support for discourse-based interventions, using books or videos (Appleton & Reddy,1996; Guajardo & Watson, 2002; Peskin & Astington, 2004).

A number of studies have investigated training for children

with autism spectrum disorder to improve the understanding

TABLE 2. Summary of theory of mind tasks with variations, performance expectations by age, and references.

Task description and variations Age expectation for typical development References

Appearance-realityBelow chance performance at 3 years,

improving to near-perfect performance

by 5 years.

Flavell, 1992;Gopnik & Astington, 1988

E shows C an object, such as a sponge painted to look likea rock, and allows C to handle it, discovering its true

nature. E asks C: (a)‘‘What is this, reallyand truly?Is it reallya sponge orreally a rock?’’ (b) ‘‘What does it looklike to youreyes? Does it look like a sponge or does it look like a rock?’’

Variation: E shows C a white object such as a fish-shaped paper cutout. Fish is placed behind a coloredfilter,removed,and replaced behindthe filter. E asks C:(a) ‘‘What color is the fish, really and truly? Is it reallywhiteor really blue?’’ (b) ‘‘What color doesit looklike toyour eyes? Does it look white or does it look blue?’’

Flavell, 1992

Unexpected contentsBelow chance performance at 3 years,

improving to near-perfect performanceby 5 years.

Astington & Jenkins, 1999E shows C a container that indicates its contents, such as aneggcarton. E asks C what is in container.C names expectedcontents. E reveals actual contents, such as pencils. Tocheck memory, E asksC whatis in box.E asksC: (a) ‘‘Whatdid you think was in here?’’ (b)‘‘ If I close this up and show itto [someone else], what will he/she think is in here?’’

Variation: Change wording of test questions. (a) ‘‘Whatdid you think was in here before I opened it?’’ (b) ‘‘If Iclose this up and show it to [someone else], what willhe/she think is in here before I open it?’’

Expect better performance using variation. Lewis & Osborne, 1990

Change of locationBelow chance performance at 3 years,

improving to near-perfect performanceby 5 years.

Miller, 2004E introduces C to puppet who likes to play with small toy. Twodissimilar containers are available. Puppet places toy incontainer A and exits. E moves toy from container A tocontainer B, replacing lids (E may encourage C to help). Tocheck memory, E asks C where the toy was placed bypuppet and where the toy is now. E asks C, ‘‘Where doesthe puppet think the toy is?’’

Variation 1: Change wording of test question. ‘‘Where willthe puppet look for the toy?’’ or ‘‘Where will the puppetlook first for the toy?’’

Expect somewhat better performancein younger children and children withlanguage impairment using Variation 1.

Miller, 2001;Siegal & Beattie, 1991

Variation 2: Present scenario as narrative, with or with-out figures or illustrations. Wimmer & Perner, 1983

Second-order false beliefBetter than chance performance starting

at 6 years, improving over next 3 to4 years.

Perner & Wimmer, 1985E presents a narrative, supported by figures or illustrations.Example: Big Bird and Elmo see a balloon seller in thepark. Elmo goes home to get money. Big Bird witnessesballoon seller moving to school playground, then goeshome. Elmo looks out window and sees balloon seller atschool playground. Later, Big Bird goes to Elmo’s houseand is told he went to buy a balloon. E asks C, ‘‘Wheredoes Big Bird think Elmo is?’’

Note. E = experimenter; C = child.




of mental states (e.g., Bell & Kirby, 2002; Ozonoff & Miller,1995; Steerneman & Huskens, 1996; Wellman et al., 2002).Usually, these studies have been focused on improving falsebelief performance or recognition of emotion and have notexamined how language change might be a cause or effect of change in theory of mind. Chin and Bernard-Opitz (2000),

however, trained 3 high-functioning children with autism onconversational skills (e.g., topic maintenance, turn taking) andmeasured their performance on false belief tasks. There wasevidence of improved conversational skills for 2 of the chil-dren, but they did not succeed in passing false belief tasks.

Hypothetical Case Study

Description of a hypothetical case may make the recom-mendations for assessment and intervention more concrete.Charlie is a boy age 6;2 whose language development is atapproximately a 3-year-old level with an MLU in morphemesof 3.8. Genetic syndromes (e.g., fragile X) have been ruled

out, but some mild autism-like signs have been noted, suchas sensitivity to loud noises, some repetitive behaviors, and alack of interest in using language for social closeness. Charliehas been receiving speech-language services for almost2 years and has made considerable progress. Two of the goalsthat are currently being emphasized are expanding his recep-tive and expressive vocabulary and increasing social uses of language.

The speech-language pathologist, Karen, decides to assessCharlie’s theory of mind development in several ways. Giventhat social use of language is an issue for Charlie, Karen wantsto determine whether he can establish joint attention with apartner. She observes his play with his mother, and in her owninteractions with him creates opportunities for him to respondto and initiate joint attention. She notes that when the adultpartner calls his name or says, ‘‘Look at this!’’ he will followher gaze to a nearby object. If the adult points to an objectacross the room or behind him, he will lookat it, even if he hasto turn around. When presented with a novel, interesting toy,he will usually initiate joint attention, looking from the toy tothe adult and back and appearing interested or happy. Karenconcludes that she does not need to focus on joint attention inintervention.

To assess understanding and use of mental state terms,Karen chooses several picture books that include characterswho are experiencing emotions such as happiness, fear, or surprise; characters wanting or wishing for objects or events;

and events in which there is a potential for false belief (e.g.,one character hiding from others, a character’s expectationsbeing violated, a character telling a lie or a joke). An exampleprovided in Cassidy et al. (1998) is Harry the Dirty Dog byGene Zion (1956), in which Harry gets so dirty that his familyfalsely believes he is not their dog. Karen reads the books withCharlie, simplifying the text when necessary to match hiscomprehension level. She finds that although Charlie talksabout what he sees in the pictures, he rarelyrefers to emotions,desires, or beliefs spontaneously. She tries prompting (e.g.,‘‘Did Harry’s family know it was him? Why not?’’) and elicitssome mention of mental states, usually accurate. However,it is clear that Charlie is more interested in the characters’

behavior than their mental states.

Because mental state terms are often used in sentencecomplement structures, Karen also assesses Charlie’s under-standing of sentence complements with both mental and com-munication verbs. She follows the method described by Haleand Tager-Flusberg (2003). One character performs an actionon another character (e.g., tickling) but says he did it to a dif-

ferent character. Karen asks Charlie, ‘‘What did he say?’’ or ‘‘Who did he say he tickled?’’ Karen conducts several trials,using different characters and actions. Charlie respondscorrectly about 25% of the time.

Karen assesses Charlie’s performance on several falsebelief tasks incorporated into play periods. For instance, shesets up a change of location scenario in which a teddy bear leaves his bouncy ball in a box rather than a basket. The teddybear is then placed outside the room (to ‘‘visit a friend’’), andKaren helps Charlie move the ball from the box to the basket.She uses different test questions to assess the effect of syn-tactic complexity on Charlie’s performance. When askedwhere the teddy bear thinks the ball is, Charlie answers cor-

rectly about 30% of the time; if he was simply guessing, heshould get about 50% correct. When Charlie is asked wherethe teddy bear will look for the ball, his performance improvesto about 70% correct, suggesting that his language abilitiesmay be constraining his false belief performance.

Based on her assessment, Karen decides to include mentalstate terms (know, think, believe, want, wish, pretend ) amongCharlie’s vocabulary goals. Use of these words will be infusedinto many therapy activities, especially book reading andpretend play, and used in both simple sentences and sentencecomplement structures. Because communication verbs mayprovide a stepping-stone to mastering mental verbs, Karenwill also make sure to model communication verbs in sen-tence complements, highlighting contrasts between what some-one says and what is true. For example, Karen and Charlie readthe story of Red Riding Hood and then act it out with puppets.Karen narrates, ‘‘The wolf says, ‘I’m your grandma,’ but that’snot true. He says he’s her grandma but really he’s the wolf!’’

Karen expects that as Charlie becomes better at usingmental state terms and understanding sentence complements,his false belief performance will also improve, and she plansto probe all these areas periodically. It is worth noting that thegoals for mental verbs and sentence complements might havebeen chosen even without taking theory of mind into consid-eration. However, in this case they were specifically selectedto contribute to the broader goal of increasing Charlie’s abilityand willingness to use language to engage in rewarding social

interactions with others. These are not the onlymethods Karenwill use to reach that goal, but they are important tools thatare informed by an understanding of how language and theoryof mind interact.

Conclusion

In their efforts to help children become fully competentcommunicators, speech-language pathologists will benefitfrom a consideration of their clients’ theory of mind abilities.Children with autism spectrum disorder are known to havedifficulty with theory of mind, and for them this should bea particular area of emphasis. However, the development of

language and theory of mind are closely intertwined from




infancy on, and children with language disorders may be atrisk for problems with theory of mind, especially when prag-matic deficits are present. Conversely, an immature theoryof mind may limit a child’s language and communicativedevelopment, even one whodoes not have an autism spectrumdisorder. By using some of the examples that have been pro-

vided here and taking advantage of the growing literature onassessment and intervention for abilities related to theory of mind, speech-language pathologists can help their young clientsenjoy communication that leads to a true meeting of minds.

Acknowledgments

Thanks to Adele Miccio and Kathryn Drager for helpfulcomments and discussions.

References

Appleton, M., & Reddy, V. (1996). Teaching three-year-olds to

pass false belief tests: A conversational approach. Social Development, 56, 275–291.

Astington, J. W. (1993). The child’s discovery of the mind .Cambridge, MA: Harvard University Press.

Astington, J. W., & Baird, J. A. (2005). Representationaldevelopment and false-belief understanding. In J. W. Astington& J. A. Baird (Eds.), Why language matters for theory of mind (pp. 163–185). New York: Oxford University Press.

Astington, J. W., & Jenkins, J. M. (1999). A longitudinal studyof the relation between language and theory-of-mind devel-opment. Developmental Psychology, 35, 1311–1320.

Astington, J. W., & Jenkins, J. M. (2000). Theory of mind andsocial behavior: Causal models tested in a longitudinal study.

Merrill-Palmer Quarterly, 46 (2), 203–220.Baldwin, D. (1995). Understanding the link between joint

attention and language. In C. Moore & P. J. Dunham (Eds.), Joint attention: Its origins and role in development (pp. 131–158). Hillsdale, NJ: Erlbaum.

Baron-Cohen, S., Tager-Flusberg, H., & Cohen, D. J. (Eds.).(2000). Understanding other minds: Perspectives from devel-opmental cognitive neuroscience (2nd ed.). Oxford, England:Oxford University Press.

Bartsch, K., & Wellman, H. (1995). Children talk about themind . New York: Oxford University Press.

Bell, K. S., & Kirby, J. R. (2002). Teaching emotion and belief as mindreading instruction for children with autism. Devel-opmental Disabilities Bulletin, 30, 16–50.

Bishop, D. V. M. (1997). Uncommon understanding: Develop-ment and disorders of language comprehension in children.Hove, United Kingdom: Psychology Press.

Bretherton, I., & Beeghly, M. (1982). Talking about internalstates: The acquisition of an explicit theory of mind. Develop-mental Psychology, 18, 906–921.

Bruinsma, Y., Koegel, R. L., & Koegel, L. K. (2004). Jointattention and children with autism: A review of the literature.

Mental Retardation and Developmental Disabilities, 10,169–175.

Call, J., & Tomasello, M. (1999). A nonverbal false belief task:The performance of children and great apes. Child Develop-ment, 70, 381–395.

Carpenter, M., Nagell, K., & Tomasello, M. (1998). Socialcognition, joint attention, and communicative competencefrom 9 to 15 months of age. Monographs of the Society for

Research in Child Development, 63(4, Serial No. 255).

Cassidy, K. W., Ball, L. V., Rourke, M. T., Werner, R. S.,Feeny, N., Chu, J. Y., et al. (1998). Theory of mind conceptsin children’s literature. Applied Psycholinguistics, 19,463–470.

Charman, T., Baron-Cohen, S., Swettenham, J., Baird, G.,Cox, A., & Drew, A. (2000). Testing joint attention, imitation,and play as infancy precursors to language and theory of mind.Cognitive Development, 15, 481–498.

Chin, H. Y., & Bernard-Opitz, V. (2000). Teaching conversa-tional skills to children with autism: Effect on the developmentof a theory of mind. Journal of Autism and Developmental

Disorders, 30, 569–583.de Villiers, J. G., & de Villiers, P. A. (2000). Linguistic

determinism and the understanding of false beliefs. In P.Mitchell & K. Riggs (Eds.), Children’s reasoning and themind (pp. 191–228). Hove, United Kingdom: PsychologyPress.

de Villiers, J. G., & Pyers, J. E. (2002). Complements tocognition: A longitudinal study of the relationship betweencomplex syntax and false-belief-understanding. Cognitive

Development, 17, 1037–1060.

Dunn, J., Brown, J., Slomkowski, C., Tesla, C., & Youngblade,L. (1991). Young children’s understanding of other people’sfeelings and beliefs: Individual differences and their ante-cedents. Child Development, 62, 1352–1366.

Dyer, J. R., Shatz, M., & Wellman, H. M. (2000). Children’sbooks as a source of mental state information. Cognitive

Development, 15, 17–37.Farmer, M. (2000). Language and social cognition in children

with specific language impairment. Journal of Child Psychol-ogy and Psychiatry, 41, 627–636.

Farrar, M. J., & Maag, L. (2002). Early language developmentand the emergence of a theory of mind. First Language, 22,197–213.

Fisher, N., Happe, F., & Dunn, J. (2005). The relationshipbetween vocabulary, grammar, and false belief task perfor-

mance in children with autistic spectrum disorders and childrenwith moderate learning difficulties. Journal of Child Psychol-ogy and Psychiatry, 46, 409–419.

Flavell, J. H. (1992). Perspectives on perspective taking. In H.Beilin & P. B. Pufall (Eds.), Piaget’s theory: Prospects and

possibilities. The Jean Piaget symposium series (pp. 107–139).Hillsdale, NJ: Erlbaum.

Flavell, J. H., Everett, B. A., Croft, K., & Flavell, E. R. (1981).Young children’s knowledge about visual perception: Further evidence for the Level 1–Level 2 distinction. Developmental

Psychology, 17, 99–103.Flavell, J. H., Flavell, E. R., Green, F. L., & Moses, L. J.

(1990). Young children’s understanding of fact beliefs versusvalue beliefs. Child Development, 61, 915–928.

Gillott, A., Furniss, F., & Walter, A. (2004). Theory of mindability in children with specific language impairment. Child

Language Teaching and Therapy, 20(1), 1–11.Gleitman, L. (1990). The structural sources of verb meanings.

Language Acquisition, 1, 3–55.Gopnik, A., & Astington, J. W. (1988). Children’s understanding

of representational change and its relation to their under-standing of false belief and the appearance-reality distinction.Child Development, 59, 26–37.

Guajardo, N. R., & Watson, A. C. (2002). Narrative discourseand theory of mind development. The Journal of Genetic

Psychology, 163, 305–325.Hale, C. M., & Tager-Flusberg, H. (2003). The influence of

language on theory of mind: A training study. DevelopmentalScience, 6 (3), 346–359.




Happe, F. (1995). The role of age and verbal ability in the theoryof mind task performance of subjects with autism. Child

Development, 66, 843–855.Hresko, W. P., Reid, D. K., & Hammill, D. D. (1981). The Test

of Early Language Development . Austin, TX: Pro-Ed.Jenkins, J., & Astington, J. (1996). Cognitive factors and family

structure associated with theory of mind development in youngchildren. Developmental Psychology, 32, 70–78.

Johnston, J. R., Miller, J., & Tallal, P. (2001). Use of cognitivestate predicates by language-impaired children. International

Journal of Language and Communication Disorders, 36, 349–370.Leslie, A. M. (1987). Pretense and representation: The origins of

‘‘theory of mind’’.’’ Psychological Review, 94, 412–426.Leslie, A. M. (1994). Pretending and believing: Issues in the

theory of ToMM. Cognition, 50, 211–238.Lewis, C., Freeman, N., Kyriakidou, C., Maridaki-Kassotaki,

K., & Berridge, D. (1996). Social influences on false belief access: Specific sibling influences or general apprenticeship?Child Development, 67, 2930–2947.

Lewis, C., & Osborne, A. (1990). Three-year-olds’ problemswith false belief: Conceptual deficit or linguistic artifact?

Child Development, 61, 1514–1519.Lohmann, H., & Tomasello, M. (2003). The role of language in

the development of false belief understanding: A trainingstudy. Child Development, 74, 1130–1144.

Miller, C. A. (2001). False belief understanding in children withspecific language impairment. Journal of Communication

Disorders, 34, 73–86.Miller, C. A. (2004). False belief and sentence complement

performance in children with specific language impairment. International Journal of Language & Communication Dis-orders, 39(2), 191–213.

Moore, C., & Dunham, P. J. (1995). Joint attention: Its originsand role in development . Hillsdale, NJ: Erlbaum.

Moore, C., Pure, K., & Furrow, D. (1990). Children’s under-standing of the modal expression of speaker certainty and

uncertainty and its relation to the development of a repre-sentational theory of mind. Child Development, 61, 722–730.

Morales, M., Mundy, P., Delgado, C., Yale, M., Messinger, D.,Neal, R., & Schwartz, H. (2000). Responding to jointattention across the 6- through 24-month age period and earlylanguage acquisition. Journal of Applied Developmental

Psychology, 21(3), 283–298.Morales, M., Mundy, P., & Rojas, J. (1998). Following the

direction of gaze and language development in 6-month-olds. Infant Behavior & Development, 21(2), 373–377.

Mundy, P., Delgado, C., Block, J., Venezia, M., Hogan, A., &Seibert, J. (2003). A manual for the abridged Early SocialCommunication Scales. Retrieved December 3, 2005, fromUniversity of Miami, Psychology Department Web site:www.psy.miami.edu/faculty/pmundy/ESCS.pdf .

Ozonoff, S., & Miller, J. N. (1995). Teaching theory of mind:A new approach to social skills training for individuals withautism. Journal of Autism and Developmental Disorders, 25,415–433.

Perner, J., Frith, U., Leslie, A. M., & Leekam, S. R. (1989).Exploration of the autistic child’s theory of mind: Knowledge,belief, and communication. Child Development, 60, 689–700.

Perner, J., Ruffman, T., & Leekam, S. R. (1994). Theory of mind is contagious: You catch it from your sibs. Child

Development, 65, 1228–1238.Perner, J., Sprung, M., Zauner, P., & Haider, H. (2003). Want

that is understood well before say that, think that, and falsebelief: A test of de Villiers’s linguistic determinism onGerman-speaking children. Child Development, 74, 179–188.

Perner, J., & Wimmer, H. (1985). ‘‘John thinks that Mary thinksthatI’’: Attribution of second-order beliefs by 5- to 10-year-old children. Journal of Experimental Child Psychology, 39,437–471.

Peskin, J., & Astington, J. W. (2004). The effects of addingmetacognitive language to story texts. Cognitive Development,19, 253–273.

Peterson, C. C. (2000). Kindred spirits: Influences of siblings’perspectives on theory of mind. Cognitive Development, 15,435–455.

Peterson, C. C., & Siegal, M. (1995). Deafness, conversation,and theory of mind. Journal of Child Psychology and

Psychiatry, 36 (3), 459–474.Repacholi, B., & Gopnik, A. (1997). Early reasoning about

desires: Evidence from 14- and 18-month-olds. Developmental Psychology, 33, 12–21.

Rollins, P. R., & Snow, C. E. (1998). Shared attention andgrammatical development in typical children and children withautism. Journal of Child Language, 25, 653–673.

Rollins, P. R., Wambacq, I., Dowell, D., Mathews, L., & Reese,P. B. (1998). An intervention technique for children with

autistic spectrum disorder: Joint attentional routines. Journalof Communication Disorders, 31, 181–193.

Ruffman, T., Slade, L., & Crowe, E. (2002). The relationbetween children’s and mothers’ mental state languageand theory-of-mind understanding. Child Development, 73,734–751.

Ruffman, T., Slade, L., Rowlandson, K., Rumsey, C., &Garnham, A. (2003). How language relates to belief, desire,and emotion understanding. Cognitive Development, 18,139–158.

Seibert, J. M., Hogan, A. E., & Mundy, P. C. (1982). Assessinginteractional competencies: The Early Social-CommunicationScales. Infant Mental Health Journal, 3(4), 244–258.

Shatz, M., Diesendruck, G., Martinez-Beck, I., & Akar, D.(2003). The influence of language and socioeconomic status

on children’s understanding of false belief. Developmental Psychology, 39, 717–729.

Shatz, M., Wellman, H. M., & Silber, S. (1983). The acquisitionof mental verbs: A systematic investigation of the firstreference to mental state. Cognition, 14, 301–321.

Shields, J., Varley, R., Broks, P., & Simpson, A. (1996). Socialcognition in developmental language disorders and high-levelautism. Developmental Medicine and Child Neurology, 38,487–495.

Siegal, M., & Beattie, K. (1991). Where to look first for children’s knowledge of false beliefs. Cognition, 38, 1–12.

Sigman, M., & Kasari, C. (1995). Joint attention across contextsin normal and autistic children. In C. Moore & P. J. Dunham(Eds.), Joint attention: Its origins and role in development (pp. 189–203). Hillsdale, NJ: Erlbaum.

Sigman, M., & Ruskin, E. (1999). Continuity and change in thesocial competence of children with autism, Down syndrome,and developmental delays. Monographs of the Society for

Research in Child Development, 64(1, Serial No. 256).Steerneman, P., & Huskens, B. (1996). The development of a

social cognition training for autistic children. Psychologie in Erziehung und Unterricht, 43, 291–301.

Tager-Flusberg, H. (2000). Language and understanding minds:Connections in autism. In S. Baron-Cohen, H. Tager-Flusberg,& D. J. Cohen (Eds.), Understanding other minds: Perspec-tives from developmental cognitive neuroscience (2nd ed.,pp. 124–149). Oxford, England: Oxford University Press.

Tager-Flusberg, H., & Joseph, R. (2001, April). What are therelationships between performance on false belief tasks and




executive functions in children with neurodevelopmentaldisorders? Paper presented at the meeting of the Society for Research in Child Development, Minneapolis, MN.

Tardif, T., Wellman, H., & Cheung, K. (2004). False belief understanding in Cantonese-speaking children. Journal of Child Language, 31, 779–800.

Tomasello, M. (1995). Joint attention as social cognition. InC. Moore & P. J. Dunham (Eds.), Joint attention: Its originsand role in development (pp. 103–130). Hillsdale, NJ:Erlbaum.

Tomasello, M., & Farrar, M. J. (1986). Joint attention and earlylanguage. Child Development, 57, 1454–1463.

van der Lely, H. K. J., Hennessey, S., & Battell, J. (1999). Linguistic determinism and false belief: Insight from childrenwith specific language impairment . Paper presented at the24th Annual Boston University Conference on LanguageDevelopment, Boston.

Wellman, H. M., Baron-Cohen, S., Caswell, R., Gomez, J. C.,Swettenham, J., Toye, E., & Lagatutta, K. (2002). Thought-bubbles help children with autism acquire an alternative to atheory of mind. Autism, 6, 343–363.

Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysisof theory-of-mind development: The truth about false belief.Child Development, 72, 655–684.

Wetherby, A. M., & Prizant, B. M. (1993). Communication and Symbolic Behavior Scales. Baltimore: Brookes.

Wetherby, A. M., & Prizant, B. M. (2002). Communication and Symbolic Behavior Scales Developmental Profile. Baltimore:Brookes.

Whalen, C., & Schreibman, L. (2003). Joint attention trainingfor children with autism using behavior modification procedures.

Journal of Child Psychology and Psychiatry, 44, 456–468.Wimmer, H., & Perner, J. (1983). Beliefs about beliefs:

Representation and constraining function of wrong beliefs in youngchildren’s understanding of deception. Cognition, 13, 103–128.

Woolfe, T., Want, S., & Siegal, M. (2002). Signposts to develop-ment: Theory of mind in deaf children. Child Development, 73,768–778.

Youngblade, L. M., & Dunn, J. (1995). Individual differences inyoung children’s pretend play with mothers and siblings: Linksto relationships and understanding of other people’s feelingsand beliefs. Child Development, 66, 1472–1492.

Ziatas, K., Durkin, K., & Pratt, C. (1998). Belief term develop-ment in children with autism, Asperger syndrome, specificlanguage impairment, and normal development: Links totheory of mind development. Journal of Child Psychologyand Psychiatry and Allied Disciplines, 39, 755–763.

Zion, G. (1956). Harry the dirty dog. New York: HarperCollins. .

Received July 14, 2005

Revision received December 8, 2005Accepted January 22, 2006DOI: 10.1044/1058-0360(2006/014)

Contact author: Carol Miller, Department of CommunicationSciences and Disorders, 110 Moore Building, The PennsylvaniaState University, University Park, PA 16802.E-mail: [email protected]


miller_06_offprint.pdf

Documents